A Framework for Efficient and Anonymous Web Usage Mining Based on Client-Side Tracking

Abstract

Web Usage Mining (WUM), a natural application of data mining techniques to the data collected from user interactions with the web, has greatly concerned both academia and industry in recent years. Through WUM, we are able to gain a better understanding of both the web and web user access patterns; a knowledge that is crucial for realization of full economic potential of the web. In this chapter, we describe a framework for WUM that particularly satisfies the challenging requirements of the web personalization applications. For on-line and anonymous web personalization to be effective, WUM must be accomplished in real-time as accurately as possible. On the other hand, the analysis tier of the WUM system should allow compromise between scalability and accuracy to be applicable to real-life web-sites with numerous visitors. Within our WUM framework, we introduce a distributed user tracking approach for accurate, efficient, and scalable collection of the usage data. We also propose a new model, the Feature Matrices (FM) model, to capture and analyze users access patterns. With FM, various features of the usage data can be captured with flexible precision so that we can trade off accuracy for scalability based on the specific application requirements. Moreover, due to low update complexity of the model, FM can adapt to user behavior changes in real-time. Finally, we define a novel similarity measure based on FM that is specifically designed for accurate classification of partial navigation patterns in real-time. Our extensive experiments with both synthetic and real data verify correctness and efficacy of our WUM framework for efficient web personalization.KeywordsAssociation RuleConcept SpaceDynamic ClusterUser SessionClient MachineThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.